Results 1 
2 of
2
Global Optimization of Statistical Functions with Simulated Annealing
 Journal of Econometrics
, 1994
"... Many statistical methods rely on numerical optimization to estimate a model’s parameters. Unfortunately, conventional algorithms sometimes fail. Even when they do converge, there is no assurance that they have found the global, rather than a local, optimum. We test a new optimization algorithm, simu ..."
Abstract

Cited by 126 (1 self)
 Add to MetaCart
Many statistical methods rely on numerical optimization to estimate a model’s parameters. Unfortunately, conventional algorithms sometimes fail. Even when they do converge, there is no assurance that they have found the global, rather than a local, optimum. We test a new optimization algorithm, simulated annealing, on four econometric problems and compare it to three common conventional algorithms. Not only can simulated annealing find the global optimum, it is also less likely to fail on difficult functions because it is a very robust algorithm. The promise of simulated annealing is demonstrated on the four econometric problems.
The Effects of Different Experimental Designs on Parameter Estimation in the Kinetics of a Reversible Chemical Reaction
, 2000
"... Using the experimental design methods, instead of deterministic ones commonly used by chemists, to estimate the kinetic constants is a very important subject in chemical researches, since suchobtained kinetic parameters have much better statistical properties. Doptimum design is a popular statist ..."
Abstract
 Add to MetaCart
Using the experimental design methods, instead of deterministic ones commonly used by chemists, to estimate the kinetic constants is a very important subject in chemical researches, since suchobtained kinetic parameters have much better statistical properties. Doptimum design is a popular statistical technique, which emphasizes to obtain the estimated parameters with the smallest content of the confidence region. But for nonlinear models, such as the kinetic model of a reversible reaction, the Doptimum design is locally optimal at the value of the initial chosen parameters. The goal of this article is try to put different experimental design techniques, i.e., uniform design, orthogonal design and Doptimum design into a common framework, and to attempt to gain some insight on when and where which of these three experimental methods can be expected to work well. The extensive Monte Carlo experiments have been done in order to compare the performance of these methods from point of vi...